add spacexr

tools/spacexr/.shed.yml

@@ -0,0 +1,13 @@
+name: spacexr
+owner: iuc
+description: Cell type identification and cell type-specific differential expression in spatial transcriptomics
+homepage_url: https://github.com/dmcable/spacexr/tree/master
+long_description: Computational methods for cell type identification (RCTD) and differential expression (C-SIDE) on spatial transcriptomics datasets
+remote_repository_url: https://github.com/galaxyproject/tools-iuc/tree/master/tools/spacexr
+categories: 
+- Spatial Omics
+- Single Cell
+suite:
+  name: "suite_spacexr"
+  description: "A suite of Galaxy tools designed to work with the spacexr-tools collection."
+  type: repository_suite_definition

tools/spacexr/macros.xml

@@ -0,0 +1,104 @@
+<macros>
+    <token name="@TOOL_VERSION@">2.2.1=r43hdfd78af_0</token>
+    <token name="@VERSION_SUFFIX@">0</token>
+    <token name="@PROFILE@">23.0</token>
+    <xml name="requirements">
+        <requirements>
+            <requirement type="package" version="@TOOL_VERSION@">r-spacexr</requirement>
+            <yield/>
+        </requirements>
+    </xml>
+    <xml name="edam">
+        <edam_topics>
+            <edam_topic>topic_4019</edam_topic>
+            <edam_topic>topic_4028</edam_topic>
+            <edam_topic>topic_3308</edam_topic>
+        </edam_topics>
+        <edam_operations>
+            <edam_operation>operation_3223</edam_operation>
+        </edam_operations>
+    </xml>
+    <token name="CSIDE_COMMON_RUN"><![CDATA[
+#if str($type.cell_types) != '':
+cell_types = c($cell_types),
+#end if
+cell_type_threshold = $type.cell_type_threshold,
+gene_threshold = $type.gene_threshold,
+doublet_mode = $type.doublet_mode,
+#if str($type.weight_threshold) != '':
+weight_threshold = $type.weight_threshold,
+#end if
+sigma_gene = $type.sigma_gene,
+PRECISION.THRESHOLD = $type.precision_threshold,
+#if str($type.cell_types_present) != '':
+cell_types_present = c($cell_types_present),
+#end if
+fdr = $type.fdr,
+test_genes_sig = $type.test_genes_sig,
+logs = $type.logs
+    ]]></token>
+    <token name="CSIDE_SINGLE_RUN"><![CDATA[
+myRCTD <- run.CSIDE.single(myRCTD,
+                            explanatory.variable,
+                            normalize_expr = $type.normalize_expr,
+                            log_fc_thresh = $type.log_fc_thresh,
+                            fdr_method = "BH", # default
+                            medv = $type.medv,
+                            CSIDE_COMMON_RUN
+                            )
+    ]]></token>
+    <xml name="sanitizer">
+        <sanitizer invalid_char="">
+            <valid initial="string.ascii_letters,string.digits">
+                <add value="_" />
+            </valid>
+        </sanitizer>
+    </xml>
+    <xml name="patho_barcode" token_help="Comma separated barcodes of the pathological region.">
+        <param name="pathologic_barcode" type="text" optional="false" label="Barcodes" help="@HELP@">
+            <expand macro="sanitizer"/>
+        </param>
+    </xml>
+    <xml name="radius">
+        <param argument="radius" type="integer" min="0" value="50" label="Radius" help="The radius of the exponential filter. Approximately, the distance considered to be a relevant interaction."/>
+    </xml>
+    <xml name="cside_common_input">
+        <param argument="cell_types" type="text" optional="true" label="Cell types used for CSIDE" help="(Comma separated) If null, cell types will be chosen with aggregate occurences of at least 'cell type threshold'."/>
+        <param argument="cell_type_threshold" type="integer" min="0" value="125" label="Cell type threshold" help="Min occurence of number of cells for each cell type to be used."/>
+        <param argument="gene_threshold" type="float" min="0" value="0.00005" label="Gene threshold" help="Minimum average normalized expression required for selecting genes."/>
+        <param argument="doublet_mode" type="boolean" truevalue="TRUE" falsevalue="FALSE" checked="true" label="Use RCTD doublet mode weights?" help="Otherwise, uses RCTD full mode weights." />
+        <param argument="weight_threshold" type="float" min="0" value="" optional="true" label="Weight threshold" help="The threshold of total normalized weights across all cell types in 'cell types' per pixel to be included in the model."/>
+        <param argument="sigma_gene" type="boolean" truevalue="TRUE" falsevalue="FALSE" checked="true" label="Fit gene specific overdispersion parameter?" help="If FALSE, overdispersion parameter is same across all genes." />
+        <param argument="precision_threshold" type="float" min="0" value="0.05" label="Precision threshold" help="For checking for convergence, the maximum parameter change per algorithm step."/>
+        <param argument="cell_types_present" type="text" optional="true" label="Cell types present" help="(Comma separeated) cell types (a superset of 'cell types') to be considered as occuring often enough to consider for gene expression contamination during the step filtering out marker genes of other cell types."/>
+        <param argument="fdr" type="float" min="0" value="0.01" label="FDR" help="False discovery rate for hypothesis testing."/>
+        <param argument="test_genes_sig" type="boolean" truevalue="TRUE" falsevalue="FALSE" checked="true" label="Genes will be tested for significance."/>
+        <param argument="logs" type="boolean" truevalue="TRUE" falsevalue="FALSE" checked="false" label="write progress to log?"/>
+    </xml>
+    <xml name="cside_single_input">
+        <expand macro="cside_common_input"/>
+        <param argument="normalize_expr" type="boolean" truevalue="TRUE" falsevalue="FALSE" checked="false" label="Constrain total gene expression to sum to 1 in each condition?"/>
+        <param argument="log_FC_thresh" type="float" min="0" value="0.4" label="LogFC threshold" help="The natural log fold change cutoff for differential expression."/>
+        <param argument="medv" type="float" min="0" value="0.5" label="Explanatory.variable cutoff" help="For determining if enough pixels for each cell type have explanatory-variable greater than or less than this value."/>
+    </xml>
+    <xml name="output">
+        <section name="output" title="Output Options">
+            <param name="output_selector" type="select" multiple="true" optional="true" display="checkboxes" label="Select / Deselect all">
+                <option value="rds">RDS file</option>
+                <option value="rscript">R script</option>
+                <yield/>
+            </param>
+        </section>
+    </xml>
+    <xml name="citations">
+        <citations>
+            <citation type="doi">10.1038/s41587-021-00830-w</citation>
+            <citation type="doi">10.1038/s41592-022-01575-3</citation>
+            <citation type="bibtex">@Manual{github,
+                title = {SpatialeXpressionR: Cell type identification and cell type-specific differential expression in spatial transcriptomics.},
+                author = {Dylan Cable},
+                url = {https://github.com/dmcable/spacexr}}
+            </citation>
+        </citations>
+    </xml>
+</macros>

tools/spacexr/spacexr_cside.xml

@@ -0,0 +1,233 @@
+<tool id="spacexr_cside" name="CSIDE" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
+    <description>Cell type-specific differential expression with C-SIDE</description>
+    <macros>
+        <import>macros.xml</import>
+    </macros>
+    <expand macro="edam"/>
+    <expand macro="requirements">
+        <requirement type="package" version="9.0">openssh</requirement>
+    </expand>
+    <command detect_errors="exit_code"><![CDATA[
+        export GALAXY_SLOTS=2 &&
+        mkdir -p 'inputs' 'results' 'logs' 'figures' &&
+        ln -s '$rctd' 'inputs/rctd.rds' &&
+        touch 'results/cside_script.R' &&
+        cat '$cside_script' > 'results/cside_script.R' &&
+        Rscript '$cside_script'
+        #if 'plots' in $output_selector:
+            mv 'results/de_plots/*.pdf' 'results/de_plots_quant/*.pdf' 'results/de_plots_two_regions/*.pdf' 'figures'
+        #end if
+    ]]></command>
+    <configfiles>
+        <configfile name="cside_script"><![CDATA[
+#if $type.cell_types != '':
+    #set $cell_types_list = ['"' + str(x.strip()) + '"' for x in str($type.cell_types).split(',')]
+    #set $cell_types = ','.join($cell_types_list)
+#end if
+#if str($type.cell_types_present) != '':
+    #set $cell_types_present_list = ['"' + str(x.strip()) + '"' for x in str($type.cell_types_present).split(',')]
+    #set $cell_types_present = ','.join($cell_types_present_list)
+#end if
+#if $type.de_type == 'point_density':
+    #set $pathologic_barcode_list = ['"' + str(x.strip()) + '"' for x in str($type.pathologic_barcode).split(',')]
+    #set $pathologic_barcode = ','.join($pathologic_barcode_list)
+#end if
+#if $type.de_type == 'custom':
+    #set $pathologic_barcode_list = ['"' + str(x.strip()) + '"' for x in str($type.pathologic_barcode).split(',')]
+    #set $region_list = [','.join($pathologic_barcode_list)]
+    #for $i, $region in enumerate($type.region):
+        #if str($region.next_pathologic_barcode) != '':
+            #set $next_pathologic_barcode_list = ['"' + str(x.strip()) + '"' for x in str($type.next_pathologic_barcode).split(',')]
+            #set $next_pathologic_barcode = [','.join($next_pathologic_barcode_list)]
+            $region_list.append($next_pathologic_barcode)
+        #end if
+    #end for
+#end if
+
+# cside script
+# This file is used to specify the parameters for the cside from spacexr package
+
+# Load the spacexr library
+library('spacexr')
+library('Matrix')
+library('doParallel')
+
+# load RCTD object
+myRCTD <- readRDS('inputs/rctd.rds')
+core <- Sys.getenv("GALAXY_SLOTS")
+# set core
+
+myRCTD@config[["max_cores"]] <- core
+
+# CSIDE
+#if str($type.de_type) == 'non_parametric':
+myRCTD <- run.CSIDE.nonparam(myRCTD,
+                            df = $type.df,
+                            barcodes = NULL, # use all barcodes
+                            CSIDE_COMMON_RUN
+                            )
+
+#else if str($type.de_type) == 'point_density':
+pathogen_coords <- myRCTD@spatialRNA@coords[c($pathologic_barcode),]
+barcodes <- colnames(myRCTD@spatialRNA@counts)
+explanatory.variable <- exvar.point.density(myRCTD,
+                                            barcodes,
+                                            pathogen_coords,
+                                            radius = $type.radius
+                                            )
+CSIDE_SINGLE_RUN
+
+
+#else if str($type.de_type) == 'cell2cell':
+barcodes <- colnames(myRCTD@spatialRNA@counts)
+explanatory.variable <- exvar.celltocell.interactions(myRCTD,
+                                                        barcodes,
+                                                        '$type.cell_type',
+                                                        radius = $type.radius
+                                                        )
+CSIDE_SINGLE_RUN
+
+
+#else if str($type.de_type) == 'XY':
+    #if str($type.xy) == 'X':
+explanatory.variable <- as.integer(myRCTD@spatialRNA@coords$x &gt; $type.lim)
+names(explanatory.variable) <- rownames(myRCTD@spatialRNA@coords)
+
+    #else
+explanatory.variable <- as.integer(myRCTD@spatialRNA@coords$y &gt; $type.lim)
+names(explanatory.variable) <- rownames(myRCTD@spatialRNA@coords)
+    #end if
+CSIDE_SINGLE_RUN
+
+#else:
+myRCTD <- run.CSIDE.regions(myRCTD,
+                            $region_list,
+                            log_fc_thresh = $type.log_fc_thresh,
+                            CSIDE_COMMON_RUN
+                            )
+#end if
+
+
+# save the results
+
+# save significant genes in each cell type
+cell_types <- names(myRCTD@de_results[["sig_gene_list"]])
+for (cell_type in cell_types) {
+    df <- myRCTD@de_results[["sig_gene_list"]][[cell_type]]
+    assign(cell_type, df)
+    write.table(df, file = paste0("results/", cell_type, "_sig.tabular"), sep = "\t", quote = FALSE)
+}
+# save all genes in each cell type
+cell_types <- names(myRCTD@de_results[["all_gene_list"]])
+for (cell_type in cell_types) {
+    df <- myRCTD@de_results[["all_gene_list"]][[cell_type]]
+    assign(cell_type, df)
+    write.table(df, file = paste0("results/", cell_type, ".tabular"), sep = "\t", quote = FALSE)
+}
+
+# create plots
+#if 'plots' in $output_selector:
+make_all_de_plots(myRCTD, "figures")
+#end if
+# save rds file
+#if 'rds' in $output_selector:
+saveRDS(myRCTD, file = 'results/cside_results.rds')
+#end if
+        ]]></configfile>
+    </configfiles>
+    <inputs>
+        <param name="rctd" type="data" format="rds" label="RCTD object" help="annotated RCTD object"/>
+        <conditional name="type">
+            <param name="de_type" type="select" label="Type of covariates for explaining differential expression with C-SIDE">
+                <option value="non_parametric">Smooth spatial pattern (non-non_parametric)</option>
+                <option value="point_density">Proximity to pathology</option>
+                <option value="cell2cell">Cell-to-cell interaction</option>
+                <option value="XY">define X or Y axis</option>
+                <option value="custom">Custom spatial locations</option>
+            </param>
+            <when value="non_parametric">
+                <param argument="df" type="integer" min="0" value="15" label="Degrees of freedom" help="The degrees of freedom, or number of basis functions to be used in the model."/>
+                <expand macro="cside_common_input"/>
+            </when>
+            <when value="point_density">
+                <expand macro="patho_barcode"/>
+                <expand macro="radius"/>
+                <expand macro="cside_single_input"/>
+            </when>
+            <when value="cell2cell">
+                <param name="cell_type" type="text" optional="false" label="Cell type for which to compute density">
+                    <expand macro="sanitizer"/>
+                </param>
+                <expand macro="radius"/>
+                <expand macro="cside_single_input"/>
+            </when>
+            <when value="XY">
+                <param name="lim" type="integer" value="" optional="false" label="Axis" help="The number on X or Y axis to discriminate two spatial regions"/>
+                <param name="xy" type="boolean" truevalue="X" falsevalue="Y" checked="true" label="Is the number on X axis?"/>
+                <expand macro="cside_single_input"/>
+            </when>
+            <when value="custom">
+                <expand macro="patho_barcode" help="Comma separated barcodes of the custom region."/>
+                <repeat name="region" min="0" title="Next custom region">
+                    <param name="next_pathologic_barcode" type="text" optional="false" label="Barcodes" help="Comma separated barcodes of the custom region.">
+                        <expand macro="sanitizer"/>
+                    </param>
+                </repeat>
+                <expand macro="cside_common_input"/>
+                <param argument="log_FC_thresh" type="float" min="0" value="0.4" label="logFC cutoff for differential expression"/>
+            </when>
+        </conditional>
+        <expand macro="output">
+            <option value="plots">DEG plots</option>
+        </expand>
+    </inputs>
+    <outputs>
+        <collection name="de_results" type="list" label="${tool.name} on ${on_string}: DE Results">
+            <discover_datasets pattern="(?P&lt;name&gt;.+)\.tabular$" format="tabular" directory="results"/>
+        </collection>
+        <collection name="de_plots" type="list" label="${tool.name} on ${on_string}: DE plots">
+            <discover_datasets pattern="(?P&lt;name&gt;.+)\.pdf$" format="pdf" directory="figures"/>
+            <filter>output['output_selector'] and 'plots' in output['output_selector']</filter>
+        </collection>
+        <data name="out_rds" format="rds" from_work_dir="results/cside_results.rds" label="${tool.name} on ${on_string}: RDS file">
+            <filter>output['output_selector'] and 'rds' in output['output_selector']</filter>
+        </data>
+        <data name="out_rscript" format="txt" from_work_dir="results/cside_script.R" label="${tool.name} on ${on_string}: RScript">
+            <filter>output['output_selector'] and 'rscript' in output['output_selector']</filter>
+        </data>
+    </outputs>
+    <tests>
+        <test expect_num_outputs="2">
+            <param name="rctd" value="myRCTD_merfish.rds"/>
+            <param name="de_type" value="non_parametric"/>
+            <param name="cell_types" value="Astrocytes"/>
+            <param name="cell_type_threshold" value="10"/>
+            <param name="gene_threshold" value="0.001"/>
+            <param name="fdr" value="0.25"/>
+            <param name="output_selector" value="rscript" />
+            <output_collection name="de_results" type="list">
+                <element name="type8" ftype="tabular">
+                    <assert_contents>
+                        <has_text_matching expression="Hello"/>
+                    </assert_contents>
+                </element>
+                <element name="type8_sig" ftype="tabular">
+                    <assert_contents>
+                        <has_text_matching expression="Hello"/>
+                    </assert_contents>
+                </element>
+            </output_collection>
+            <output name="out_rscript">
+                <assert_contents>
+                    <has_text_matching expression="run.CSIDE.nonparam"/>
+                </assert_contents>
+            </output>
+        </test>
+    </tests>
+    <help><![CDATA[
+
+Cell type-Specific Inference of Differential Expression, or CSIDE, is part of the spacexr R package for learning cell type-specific differential expression from spatial transcriptomics data.
+
+    ]]></help>
+    <expand macro="citations" />
+</tool>